keytest.py now only prints one time the command that it is executing for
an assert. The command is already printed in the function responsible
for executing it, so it is not necessary to print the command before
calling that function.
* some functionality is in new modules now
new header locations and library names: QtConcurrent and QtWidgets
* method setResizeMode is renamed to setSectionResizeMode
* deprecated QAbstractItemModel::reset() is dropped now
* platform specific code like QApplication::syncX() is not common anymore
* QString::fromAscii() is dropped now
* some QDesktopServices methods has been moved to QStandardPaths
This test should fail because #7673 is not yet fixed in trunk.
Note that it is fixed in branch.
This is the first autotest that uses an existing .lyx file,
bug-7673.lyx.
If the test opens an existing .lyx file and crashes, it will leave
around a .emergency file. If a test with that same name is run again,
LyX will try to recover the .emergency file, which could throw
off the test.
This is implemented for both CMake and autotools.
LyX 2.0.x supports currently 147 LaTeX-packages. When LyX is first installed on a PC it took therefore about 10 minutes to install all supported and required packages. With the new installation method the installation speed only depends on the Internet speed and lasts with an average bandwidth now only 3.5 minutes.
(chkconfig.ltx asks LaTeX for each package separately and after each package installation the file name database of LaTeX was rebuild. And that cost a lot of time/CPU power)
To reproduce some bugs, tests need to open an existing LyX file. This
is now possible with CMake and is consistent with testing under
autotools in the sense that the .lyx file is referred to in the test
by being in the parent directory.
If the test is named bug-XXXX-in.txt and there exists a file
bug-XXXX.lyx, that file will be copied to the parent directory of
out-home in the build directory.
I do not see where this is really useful (contrary to real plaintext). And it breaks spell checking with hunspell and 8bit disctionaries (part of bug #8526)
